Courier by Courier

# Fix list: Courier

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/courier, the October 2026 research run, assessed 1 October 2026. Grade BB, 70.5 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Courier: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). $0.005 a send on Business, published without login (20). Developer plan free for 10,000 sends a month, but the pricing page doesn't say whether a card is needed (15). A person signs up in the browser and copies a key (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Security & auth, 56 out of 100, up to 7.7 more on the total

Why it scored 56: Plain `pk_` API keys scoped to a Test or Production environment, rotatable in settings, with no scopes. The hosted MCP takes the key in an `api_key` header and has no OAuth (20). A Test key can't send real notifications, but there's no read-only key, and the MCP server includes `delete_profile`, `delete_list`, `delete_tenant` and `delete_provider` (5). Tools return profiles, templates and message logs that the operator's own systems wrote, with no prompt-injection guidance (7). Message logs per send, kept 7 days on Developer, 30 on Business and a year on Enterprise, plus an audit trail of key, user and template changes on Enterprise only (12). Trust centre at security.courier.com lists SOC 2, HIPAA and PCI DSS and a 2026 pen test on request, and names "Responsible Disclosure" without details. No security.txt per the 30 September check, and the SDK's SECURITY.md points to Stainless (12).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 3. Agent ergonomics, 68 out of 100, up to 5.2 more on the total

Why it scored 68: 170 MCP tools in 13 groups with no toolsets, filtering or read-only subset (5). List endpoints return a `paging` cursor (3). Cursor pagination and filters such as templates by tag (17). Error bodies carry a `message` and `type`, with rate-limit headers on 429, but no documented codes to branch on (12). `Idempotency-Key` on writes, scoped per endpoint, 25 hours by default and up to a year. The docs mark write tools, and we found no readOnlyHint or destructiveHint (16). A send needs only a recipient and content, and official SDKs cover Node and Python among others (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 4. Transparency & trust, 65 out of 100, up to 3.1 more on the total

Made of editorial 40, provenance 90.

Why it scored 65: Closed service with published terms, the SDKs under Apache-2.0 (15). The privacy policy, last updated 8 December 2025, names trycourier.com, Inc. and the EU-U.S. Data Privacy Framework, but gives no retention period, while the pricing page gives log retention per plan. No DPA or subprocessor list found (14). No deprecation policy. The only notice we found is the archived open-source MCP repo, whose registry entry now points at the hosted server (5). The trust centre names AWS and Google Cloud with no regions and no subprocessor list (6).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- security.txt: not found (0 of 10)

## 5. Schema & documentation, 84 out of 100, up to 2.6 more on the total

Why it scored 84: Public OpenAPI file at /docs/openapi-specs/openapi.documented.yml (25). llms.txt with about 230 entries, each a Markdown page (10). The API overview says what each limit and header is for, and the MCP tools page marks every write tool, but we couldn't read the hosted server's own tool definitions (15). Typed request bodies in the OpenAPI file and the Stainless SDKs, with message content as a flexible object (12). Examples throughout, but errors are documented only as a `message` and `type` body and the status classes, with no list of error types (10). The SDKs follow semver with a generated CHANGELOG and the product changelog is dated, but the REST API has no version scheme (12).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 6. Reliability, 92 out of 100, up to 1.6 more on the total

Why it scored 92: Statuspage at status.courier.com with 28 incidents back to August 2024 (20). Nothing posted since the 18 June 2026 send pipeline incident, when sends sat queued for about two hours, so the last 90 days are clean (30). Limits are published with numbers. `POST /send` isn't limited by request count, list replacement is 60 a minute, preferences 50 a second and resend one per 2 seconds per message (15). Over-limit calls get 429 with `X-RateLimit-Limit` and `X-RateLimit-Remaining`, and `Idempotency-Key` replays a response for 25 hours by default or up to a year, but there's no Retry-After or backoff guidance (12). The pricing page names an Enterprise SLA without publishing its terms (5). The send API is generally available (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 7. Maintenance & community, 86 out of 100, up to 1.2 more on the total

Why it scored 86: courier-node v9.11.0 on 24 September 2026 (30). 31 SDK tags since 3 July and product changelog entries on 9 and 18 July, 18 August and 1 September (20). A closed service with a dated changelog and email support on the free plan, chat on Business (11). Current official SDKs, and the hosted server is in the official MCP registry as io.github.trycourier/courier (15). The Node SDK runs lint, build and tests in CI with release-please (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- Whether the Developer plan needs a card at signup, which the pricing page doesn't say
- Whether failed or bounced sends are billed
- The terms of the Enterprise SLA, which aren't published
- Whether the hosted MCP tools carry readOnlyHint or destructiveHint annotations

## Weaknesses

- 170 MCP tools with no toolsets, filtering or read-only mode
- MCP authentication is a raw API key header with no OAuth or scopes
- Error responses documented only as `message` and `type`, with no list of codes
- Audit trail of key and template changes is Enterprise only
- No subprocessor list, data regions or retention periods in the privacy policy

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Send an `Idempotency-Key` header on every send so a retry doesn't notify twice
- Set `routing.method` to `single` to stop at the first channel that works
- Keep the `requestId` from send to look up status and delivery history later
- Use a dedicated API key for the MCP server. Every one of the 170 tools runs with that key's full rights
- Space resends at least 2 seconds apart per message or you get a 429

## What the review panel asked for

- State if signup needs a card
- written deprecation policy

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.